Intrascope

The pitch here is narrow and honest: Intrascope does not sell you a model. It sells the layer above the models — who can reach which one, what context they get, and what it costs. If you already buy from three or four vendors, that layer is the missing piece, and the analytics view by user, model, project and vendor is the fastest way to find out where the money actually goes. Pick it when governance, not capability, is your bottleneck. An agency with per-client briefs, a 25-person startup replacing a pile of Plus seats, or a finance team that wants spend caps before a runaway agentic session — all fit. The Manifests model is the quiet winner: shared context outlives the employee who wrote it, which is more than a personal ChatGPT account will ever give you. Pass if your work lives inside Slack, Notion or Jira and you expect the AI to meet you there. Outside n8n, native integrations are not the story here, and teams expecting deep app hooks will be frustrated. Likewise skip it if you need fine-tuning or model training — this is a control plane, not an ML platform. The pricing structure asks you to do arithmetic. Seat cost is predictable; token cost is not, and the 1.3x managed multiplier means managed usage on the same million tokens that costs $10 at provider rates lands at $13. In practice, most companies we would advise start with BYOK for the vendors they already pay and use managed usage selectively. Remember the workspace fee buys no tokens at all. Two operational caveats. Power-ups exist for extra users, projects and storage, but their pricing only appears inside workspace billing, so budget a cushion above the headline tier. And the vendor has announced a price increase effective 20 October 2026 — if an annual contract is on the table, signing before then